This study empirically investigates the causal relationship between construction production and GDP for Turkey during 2005Q1-2013Q4 period. Because it is found that, there is no cointegration which means there is no long run relationship between variables, VAR Granger Causality Method is used to test the causality in short run. The findings reveal that, the causality runs from GDP to Building Production and Building Production to Non-Building Production (i.e. bidirectional relationship). Findings of this paper suggest that, because there is no long run relationship between Construction Production (Building and Non-Building) and GDP and also in short run the causality runs from GDP to Construction Production, the growth strategy based on mainly Construction Sector growth is not a good idea for Turkey.